Matrix Library Implementation Summary

Overview

The project revolved around the construction of a dedicated library, named matrix.h, for performing operations on numerical matrices using the C programming language. Matrices hold a fundamental position in programming, serving functions from representing data in tabular form to being core elements in computational tasks, including neural networks.

Installation

git clone https://github.com/0xmisha/matrix.h.git
cd matrix.h/src
make

Requirements

• Gcc
• Make
• check.h (for tests)

Technical Specificities

Data Structure

A structure matrix_t was designed to represent matrices:

typedef struct matrix_struct {
    double** matrix;
    int rows;
    int columns;
} matrix_t;

This structure includes a double pointer to represent the matrix and two integers specifying its dimensions.

Core Functions

Several vital matrix operations were implemented:

1. Creation: create_matrix initializes a matrix with specified rows and columns.
2. Deletion: remove_matrix cleans up and de-allocates a matrix.
3. Comparison: eq_matrix determines the equality of two matrices.
4. Addition & Subtraction: sum_matrix and sub_matrix handle matrix arithmetic.
5. Scalar & Matrix Multiplication: mult_number scales a matrix, while mult_matrix manages matrix-to-matrix multiplication.
6. Transpose: transpose returns the transpose of a matrix.
7. Algebraic Complements: calc_complements computes the matrix of algebraic complements.
8. Determinant: determinant evaluates the determinant of a matrix.
9. Matrix Inversion: inverse_matrix finds the inverse of a matrix.

Every function, except for eq_matrix, returns specific codes indicating successful operation or a particular error type.

Development Guidelines

• Language & Compiler: The library adheres to the C11 standard and is compiled using gcc.
• Code Location: The source code resides in the src folder on the develop branch.
• Coding Standards: Legacy constructs were avoided, ensuring compliance with the POSIX.1-2017 standard. The Google style was followed throughout development.
• Library Type: The matrix library is static, complemented by the matrix.h header file.
• Function Naming: Each function is prefixed to maintain naming clarity.
• Unit Testing: Comprehensive unit-tests were designed using the Check library, targeting at least 80% coverage per function. gcov was employed to monitor this metric.
• Compilation & Testing: A Makefile, complete with targets (all, clean, test, matrix.a, gcov_report), facilitates building and testing operations. Specifically, gcov_report generates an HTML-based coverage report.
• Precision: The library guarantees accuracy up to six decimal places for fractional components.

This endeavor successfully distilled the essence of matrix operations, fortified by rigorous development standards, into a robust C library.